Module cogsgpt.cogsmodel.nlp.text_analysis

Classes

class BaseAnalysisModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class BaseAnalysisModel(BaseModel, abc.ABC):
    def __init__(self) -> None:
        super().__init__()
        
        COGS_KEY = os.environ['COGS_KEY']
        COGS_ENDPOINT = os.environ['COGS_ENDPOINT']
        ta_credential = AzureKeyCredential(COGS_KEY)
        self.text_analytics_client = TextAnalyticsClient(
            endpoint=COGS_ENDPOINT,
            credential=ta_credential
        )

        self.supported_language = {
            LanguageType.English.value: "en",
            LanguageType.Chinese.value: "zh-hans",
        }
    
    @abc.abstractmethod
    def _analyze(self, text: str, language: str = "en") -> List:
        pass

    def run(self, *args, **kwargs) -> str:
        text = kwargs[ArgsType.TEXT.value]
        language = kwargs.get(ArgsType.SRC_LANGUAGE.value, LanguageType.English.value)
        language = self.supported_language[language]
        return str(self._analyze(text, language))

Ancestors

• BaseModel
• abc.ABC

Subclasses

• EntityLinkingModel
• EntityRecognitionModel
• KeyPhraseModel
• LanguageDetectionModel
• PIIModel
• SentimentAnalysisModel

Methods

def run(self, *args, **kwargs) ‑> str

Expand source code

def run(self, *args, **kwargs) -> str:
    text = kwargs[ArgsType.TEXT.value]
    language = kwargs.get(ArgsType.SRC_LANGUAGE.value, LanguageType.English.value)
    language = self.supported_language[language]
    return str(self._analyze(text, language))

class EntityLinkingModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class EntityLinkingModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.recognize_linked_entities(documents=[text], language=language)
        return [{
            "entity": entity.name,
            "url": entity.url,
            "data_source": entity.data_source,
            "matches": [{
                "text": match.text,
                "confidence_score": match.confidence_score
            } for match in entity.matches]
        } for entity in response[0].entities]

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC

class EntityRecognitionModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class EntityRecognitionModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.recognize_entities(documents=[text], language=language)
        return [{
            "entity": entity.text,
            "category": entity.category,
            "subcategory": entity.subcategory,
            "confidence_score": entity.confidence_score
        } for entity in response[0].entities]

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC

class KeyPhraseModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class KeyPhraseModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.extract_key_phrases(documents=[text], language=language)
        return response[0].key_phrases

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC

class LanguageDetectionModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class LanguageDetectionModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.detect_language(documents=[text])
        return [response[0].primary_language.name]

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC

class PIIModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class PIIModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.recognize_pii_entities(documents=[text], language=language)
        return [{
            "entity": entity.text,
            "category": entity.category,
            "confidence_score": entity.confidence_score
        } for entity in response[0].entities]

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC

class SentimentAnalysisModel

Helper class that provides a standard way to create an ABC using inheritance.

Expand source code

class SentimentAnalysisModel(BaseAnalysisModel):
    def _analyze(self, text: str, language: str = "en") -> List:
        response = self.text_analytics_client.analyze_sentiment(documents=[text], language=language)
        return [{
            "sentence": sentence.text,
            "sentiment": sentence.sentiment,
        } for sentence in response[0].sentences]

Ancestors

• BaseAnalysisModel
• BaseModel
• abc.ABC